Category: Indico Event
-
Model/dataset compression for optimizing the efficiency of deep networks
Wednesday 17 May/23 11:00 -12:00 (Europe/Zurich) Model/dataset compression for optimizing the efficiency of deep networks Despite the unprecedented success of deep networks in recent years, the massive size of these models and the enormous data required for training them increase the burden on storage, transmission and model training as the latter requires enormous computational costs.…
-
John Fitzgerald, “Overcoming the challenges of power transmission in the age of renewables”
Wednesday 26 Apr/23 11:00 -12:00 (Europe/Zurich) John Fitzgerald, “Overcoming the challenges of power transmission in the age of renewables” BROKEN IMAGE NOT MOVED (drupal url : https://iml.web.cern.ch//node/320) This presentation will focus on the role of superconducting technology in enabling the renewable energy transition away from fossil fuels. This is relevant as existing grids and grid…
-
IML Machine Learning Working Group
Tuesday 18 Apr/23 15:00 -18:00 (Europe/Zurich) IML Machine Learning Working Group Topic: pflow
-
[POSTPONED] IML Machine Learning Working Group
Tuesday 4 Apr/23 15:00 -18:00 (Europe/Zurich) [POSTPONED] IML Machine Learning Working Group Open topic
-
The Promise of Regenerative Medicine and AI
Monday 27 Mar/23 16:30 -17:30 (Europe/Zurich) The Promise of Regenerative Medicine and AI Today, treatments based on laboratory-cultured cells are curing cancer patients. Besides cancer, cell-based therapies have the potential to heal and rejuvenate patients suffering from diseases like Parkinson’s, Alzheimer’s, diabetes and heart disease. In parallel, the technology behind artificial neural networks has made…
-
Hyperparameter Optimization for Deep Learning Models Using High Performance Computing
Wednesday 5 Apr/23 11:00 -12:00 (Europe/Zurich) Hyperparameter Optimization for Deep Learning Models Using High Performance Computing In the past decade, Machine Learning (ML), and in particular Deep Learning (DL), has outperformed traditional rule-based algorithms on a wide variety of tasks, such as for instance image recognition, object detection and natural language processing. In CoE RAISE,…
-
Discovering fraud: can CERN tools accelerate market surveillance
Thursday 29 Jun/23 11:00 -12:00 (Europe/Zurich) Discovering fraud: can CERN tools accelerate market surveillance High Energy Physics Tools in Limit Order Book Analysis Project: A Unique cooperation between CERN and academic financial world and financial regulators. About four years ago researchers from Wageningen University (WUR)/Maastricht University (UM), the Commodity Risk Management Expertise Center (CORMEC) and…
-
ATLAS GNN Flavour Tagging
Wednesday 1 Mar/23 11:00 -12:00 (Europe/Zurich) ATLAS GNN Flavour Tagging Flavour-tagging, the identification of jets originating from b and c quarks, is a critical component of the physics programme of the ATLAS experiment. Current flavour-tagging algorithms rely on the outputs of “low-level” taggers, which are a mixture of manually optimised, physically informed algorithms and machine…
-
IML Machine Learning Working Group
Tuesday 14 Feb/23 15:00 -18:00 (Europe/Zurich) IML Machine Learning Working Group Open topic
-
Turbo: A Physical-Minded Approach to Generalized Autoencoders
Wednesday 15 Feb/23 11:00 -12:00 (Europe/Zurich) Turbo: A Physical-Minded Approach to Generalized Autoencoders This presentation explores the interconnection between the information bottleneck and a new framework called Turbo.We will first formulate a variational approximation of the information bottleneck and show how several existing models can be seen as particular cases. We then address the limitations…